Monocular vision mapping and positioning method, device, storage medium and mobile device

Abstract

The disclosure relates to a monocular vision mapping and positioning method, device, storage medium and mobile device. The method includes: respectively acquiring a first depth map and a first optical flow feature corresponding to the monocular image of the current frame, and a second depth map corresponding to the monocular image of the previous frame of the current frame; According to the first RGB image corresponding to the target image, determine the first frame point cloud; according to the second depth image and the second RGB image corresponding to the previous frame monocular image, determine the second frame point cloud; obtain the inertial measurement device in the previous frame The data measured during the time period between the first moment corresponding to the monocular image and the second moment corresponding to the current frame monocular image, and the estimated state of the inertial measurement device at the first moment; according to the data, state, first The optical flow feature determines the first movement trajectory of the mobile device within a time period; generates a dense map based on the first movement trajectory, the first frame point cloud and the second frame point cloud. In this way, the positioning accuracy of the mobile device is improved.

Description

technical field [0001] The present disclosure relates to the field of computer vision, and in particular, to a monocular vision mapping and positioning method, device, storage medium and mobile device. Background technique [0002] In many fields such as robot guidance, unmanned driving, and augmented reality (Augmented Reality, AR), environmental maps are required. In related technologies, the environmental maps mostly use monocular cameras and low-cost inertial measurement units (Inertialmeasurement unit , IMU) synchronous positioning and mapping technology (Simultaneous Localization and Mapping, SLAM) technology). Specifically, a precise odometer system is constructed by calibrating the internal and external parameters of the monocular camera and IMU, and the time synchronization between the monocular camera and IMU. In this system, firstly, the adjacent images are matched in a series of images collected by the monocular camera, and then the IMU data measured by the IMU ...

AI Technical Summary

Problems solved by technology

In addition, if the image collected by the monocular camera is lost or the IMU data is noisy, the positioning accuracy of the above method will also decrease.

Images